EP0766461A2

EP0766461A2 - Caption detection

Info

Publication number: EP0766461A2
Application number: EP96202639A
Authority: EP
Inventors: Eiji Internationaal Octrooibureau B.V. Fujisaki
Original assignee: Koninklijke Philips Electronics NV; Philips Electronics NV
Current assignee: Koninklijke Philips NV
Priority date: 1995-09-26
Filing date: 1996-09-20
Publication date: 1997-04-02
Anticipated expiration: 2016-09-20
Also published as: EP0766461A3; JPH0993504A; MY132603A; DE69609684D1; EP0766461B1

Abstract

It is an object of the invention to provide a caption moving device for realizing a high quality image expansion process so as to move only captions.

A Y/C separating 2 generates a luminance signal corresponding to an input image signal. A rate of change detecting part 3 detects a rate of change of the luminance signal and generates a rate of change signal corresponding to the rate of change. A rate of change judging part 5 compares a level of the rate of change signal with a level of a reference signal inputted separately. When the rate of change signal level exceeds the reference level, a control signal which represents valid is generated. A gate 7 outputs the luminance signal from the Y/C separating part 2 which is stored in a first memory 4 for a certain period only if the valid control signal is inputted thereto. A caption judging part 8 compares the luminance signal from the gate 7 with a reference level. When the luminance signal exceeds the reference level, a caption signal which represents valid is generated. This caption signal is written into and read from a second memory under the control of a control signal from a control signal generating part 10.

Description

The present invention relates to a method and a device for detecting captions, which are missed from a screen, (into an image region) when an image made for a television with an aspect ratio of 4:3, and comprising captions, is displayed as expanded on a television with an aspect ratio of 16:9 .
Generally, a 16:9 television is provided with a function for expanding the image to the whole of the screen. But, when this image is merely performed as expanded, any caption lines are missed from the screen. For preventing the captions from missing when the image is expanded, i.e. for realizing an expanded image in which the captions are also displayed, a caption moving device, is used in a prior art. The operating function will be explained as follows.
The image information is examined as to whether this information represents a caption or an image.
Generally, captions are white (an achromatic color). As regards a TV signal, a chroma (color) signal is not included in an image signal which represents white but this image signal consists of only a high level luminance signal. A caption judging part is aimed at this point, which is provided with a predetermined reference level for judging whether the inputted image signal represents a caption or not. Namely, the caption judging part judges that only the inputted image information of which a luminance level exceeds the reference level, i.e. only the image information of a high luminance level, represents a caption.
As the above description, an image of a high luminance level is also judged captions in the conventional 16:9 television. As a result of this, the screen during expanding is very distinct because such moved image and an image present in the caption insertion region are overlapped.
In view of the above problem, it is an object of the invention to provide a caption detection method and device for moving only captions so as to solve the distinctness of the screen during expanding an image. A further aspect of the invention is aimed that edging whose color is usually black is applied to a caption of, for example, the telecine and a sharp change of luminance level is occurred between a white (a high luminance) caption and the black (a low luminance) edging.
To this end, a first aspect of the invention provides a captions detection devices as defined in claim 1. A second aspect of the invention provides a caption detection method as defined in claim 3. A third aspect of the invention provides a video display device as defined in claim 5. Advantageous embodiments are defined in the dependent claims.
Embodiments of the present invention will now be described with reference to the accompanying drawings, in which:

Fig. 1 is a block diagram of an embodiment of a caption moving device according to the invention;
Fig. 2 is a schematic diagram of a caption signal outputted from a caption judging part to be displayed;
Fig. 3 is a picture of a 16:9 television which shows an image of a telecine for a 4:3 television comprising captions;
Fig. 4 is a picture of a 16:9 television in the case that the image shown in Fig. 3 is expanded by an expanding function which the 16:9 television comprises without performing a caption moving process;
Fig. 5 is a picture of a 16:9 television in that the case the image shown in Fig. 3 is expanded by an expanding function which the 16:9 television comprises with performing a caption moving process;
Fig. 6 is a block diagram of an embodiment of a conventional caption moving device;
Fig. 7 is an example of an image signal inputted to a caption moving device; and
Fig. 8 is a schematic diagram of a part of a general caption, around which edging is applied, to be displayed.

When an image of, for example, a telecine made for a 4:3 television comprising captions is displayed on a 16:9 television, a caption line shown by ○ is displayed in an image region, for example, as shown in Fig. 3. In Fig. 3, a region indicated by X is an image region in which captions are displayed and hereinafter will be referred as a caption region. Also, a region indicated by Y is an image region in which captions are inserted and hereinafter will be referred as a caption insertion region. As regards Y, it also will be the same in following figures.
Generally, the 16:9 television is provided with a function for expanding the image to the whole of the screen. But, when this image is merely performed as expanded, the caption line is missed from the screen as shown in Fig. 4.
For preventing the captions from missing when the image is expanded, i.e. for realizing an image in which the captions are displayed as shown in Fig. 5, a caption moving device, for example, as illustrated by a block diagram in Fig. 6 is used in a prior art. The operating function will be explained as follows.
In Fig. 6, an input image signal from an image input 61 is inputted to an A/D converter 62 and an input terminal 68a of an image/caption changing part 68. The A/D converter 62 performs an A/D conversion of the input image signal to generate image information corresponding to this image signal. This image information is written into a memory (MEM) 64. The memory 64 has a capacity enough to at least store a signal which constructs the caption region, for example, indicated by X in Fig. 3. A control signal generating part (CTRL) 65 is connected to the memory 64. The control signal generating part 65 generates a control signal used for controlling a writing/reading position of the image information inputted to the memory 64. Based on the control signal generated by the control signal generating part 65, the memory 64 begins to write image information at a point of time when the image signal reaches to the caption region X of a certain field, begins to read the stored image information at a point of time when the image signal reaches to the caption insertion region, for example, indicated by Y of a next field and begins to write image information again at a point of time when the image signal reaches to the caption region X of this field. Subsequently, the image information is read from the memory 64 and then inputted to a D/A converter 66 and a caption judging part 67. The D/A converter 66 performs a D/A conversion of the image information to generate an image signal corresponding to this image information. This image signal is inputted to an input terminal 68b of the image/caption changing part 68.
On the other hand, the image information inputted to the caption judging part 67 is judged whether this information represents a caption or an image. If the inputted image information is judged a caption, for example, a high level control signal is outputted from the caption judging part 67 to the image/caption changing part 68. On the other hand, if the information is judged an image, for example, a low level control signal is outputted therefrom. The caption judging part 67 will be explained in detail later.
The image/caption changing part 68 changes a switch based on the control signal from the caption judging part 67. Namely, if the high level control signal is inputted from the caption judging part 67, the image/caption changing part 68 set the switch to a position 68b-68c. As a result of this, an image signal, which represents a caption, for example, one field before, is outputted from an image signal output 69. Also, if the low level control signal is inputted therefrom, the image/caption changing part 68 set the switch to a position 68a-68c. As a result of this, an image signal, which represents a newest image, is outputted from the output 69. The image signal is displayed on the screen after the expanding process by means of a conventional image expanding device (not shown). Thus, the caption does not miss from the screen (the image region) if the image expanding function which the 16:9 television comprises is performed.
The process performed by the caption judging part 67 will be explained in more detail. Generally, captions are white (an achromatic color). As regards a TV signal, a chroma (color) signal is not included in an image signal which represents white but this image signal consists of only a high level luminance signal. The caption judging part 67 is aimed at this point, which is provided with a predetermined reference level for judging whether the inputted image signal represents a caption or not. Namely, the caption judging part 67 judges that only the inputted image information of which a luminance level exceeds the reference level, i.e. only the image information of a high luminance level, represents a caption.
Now, it is assumed that an image signal S of a telecine as shown in Fig. 7A and B is inputted to the image signal input 61. A signal part as indicated by a of the image signal S in Fig. 7 represents captions. On the other hand, signal parts of a high luminance level as indicated by b and c in Fig. 7 generally exist in the image signal. Now, it is assumed that a broken line R indicates the reference level set to the caption judging part 67. If the caption judgment of the image signal S is performed at this level, not only the caption signal part a but also the high level luminance signal parts b and c representing an image are judged captions. As a result of this, a control signal P shown in Figs. 7A and 7B is outputted from the caption judging part 67 to the image/caption changing part 68 and consequently an image signal of a high luminance level other than captions is also outputted from the image signal output 69. Namely, when a luminance level of the image signal is merely compared by the caption judging part 67, an image of a high luminance level substantially same as that of captions is also judged captions and then moved.
As the above description, an image of a high luminance level is also judged captions in the conventional 16:9 television. As a result of this, the screen during expanding is very distinct because such moved image and an image present in the caption insertion region are overlapped.
In view of the above problem, it is an object of the invention to provide a caption moving device for moving only captions so as to solve the distinctness of the screen during expanding an image. A further aspect of the invention is aimed that edging whose color is usually black is applied to a caption of, for example, the telecine and a sharp change of luminance level is occurred between a white (a high luminance) caption and the black (a low luminance) edging as shown Fig. 8.
Fig. 1 shows a block diagram of an embodiment of a caption moving device according to the invention. The reference numeral 1 designates a image signal input, which is connected to a Y/C separating part 2. The Y/C separating part 2 is connected to a rate of change detecting part 3 and a first memory (MEMi) 4. The rate of change detecting part 3 is connected to a rate of change judging part 5. The rate of change judging part 5 is connected to a reference level signal input 6 which inputs a reference signal to judge caption edging, and a gate 7. Also, the first memory 4 is connected to the gate 7. The gate 7 is connected to a caption judging part 8. The caption judging part 8 is connected to a second memory (MEMe) 9. The second memory 9 is connected to a control signal generating part (CTRL) 10 which generates a control signal for controlling a writing/reading position of a caption signal inputted to the memory 9. Also, the second memory is connected to a caption signal output 11.
An input image signal inputted to the image signal input 1 is inputted to the Y/C separating part 2. At the Y/C separating part 2, an Y/C separation corresponding to the input image signal is performed to generate a luminance signal and a chroma signal corresponding to the input image signal. Symbol Y represents a luminance and symbol C represents a color. In this embodiment, only a luminance signal Y corresponding to the input image signal is outputted from the Y/C separating part 2.
The luminance signal is inputted to the rate of change detecting part 3. This rate of change detecting part 3 is a so-called differential circuit (d/dt). By the differential process performed in this circuit, the rate of change of the luminance level of the input luminance signal is detected. As regards this rate of change, the larger the value thereof means the larger the change of the luminance level. As a result of the detection of the rate of change, a rate of change signal corresponding to the detected rate of change is generated. This rate of change signal is outputted toward the rate of change judging part 5.
The rate of change judging part 5 is a so-called comparator (COMP1), which judges whether the inputted rate of change signal represents a rate of change more than the certain constant level or not. For this judgment, a reference signal (REF) of a certain level is inputted to the rate of change judging part 5 from the input 6. Then, the rate of change judging part 5 compares the rate of change level of the rate of change signal from the rate of change detecting part 3 with the reference level. When the rate of change level exceeds the reference level, i.e. when the rate of change is larger, for example, a high level control signal which represents valid is generated. On the other hand, when the rate of change level does not exceed the reference level, i.e. when the rate of change is smaller, for example, a low level control signal which represents invalid is generated. This control signal is outputted toward a gate 7.
On the other hand, the luminance signal outputted from the Y/C separating part 2 is also written into a first memory 4. In this first memory 4, the inputted luminance signal is stored for a certain period. This certain period is a period is necessary for the rate of change signal generating process of the luminance signal outputted from the Y/C separating part 2 in the rate of change detecting part 3 and for the rate of change judging process of the rate of change signal generated by said process in the rate of change judging part 5. After being stored for this period, the luminance signal is read to output toward the gate 7.
The gate 7 performs an output control of the luminance signal from the first memory 4 in accordance with the state of the control signal from the rate of change judging part 5. Namely, for example, only if the high level control signal is received from the rate of change judging part 5, the luminance signal from the first memory 4 is outputted.
The luminance signal outputted from the gate 7 is inputted to the caption judging part 8. The caption judging part 8 is a so-called comparator (COMP2), which compares a luminance level of the input luminance signal with a predetermined reference level. Based on this comparison result, this part 8 judges whether the signal represents a caption or an image. Namely, when the luminance level of the input luminance signal exceeds the reference level at the caption judging part 8, this luminance signal is judged to represent a caption and then, for example, a high level caption signal which represents valid is generated. On the other hand, when the input luminance signal doesn't exceed the reference level, this luminance signal is judged to represent an image and then, for example, a low level caption signal which represents invalid is generated. Then, this caption signal is written into the second memory 9. At this point, the caption signal written into the second memory 9 is that it represents only a caption which doesn't include the black edge part applied around the caption and the image part, for example, as shown in Fig. 2.
The second memory 9 comprises a capacity, for example, enough to at least store the signal which constructs the caption region indicated by X in Fig. 3. As described above, the control signal generating part 10 which generates a control signal using a writing and reading position control of the inputted caption signal is connected to the memory 9. Based on the control signal generated by the control signal generating part 10, the second memory 9 begins to write a caption signal at a point of time when the image signal reaches to the caption region X of a certain field, begins to read the stored caption signal at a point of time when the image signal reaches to, for example, the caption insertion region indicated by Y of a next field, and begins to write again the caption signal at a point of time when the image signal of this field reaches to the caption region X. Thus, the caption signal is written into and read from the second memory 9. After that, the caption signal read from the second memory 9 is outputted from the caption signal output 11 and then displayed on the screen after performing a conventional insertion process into the image part and a conventional expansion process by an image expanding device (not shown).
Thus, an image signal (a caption signal) which represents an image of a high luminance level is not outputted from the caption signal output 11 but a caption signal which represents only a caption, for example, prior to one field, from which the edge part is removed, is outputted to display together with a newest image signal on the screen.
As described above, according to the caption moving device of the present invention, not only an image signal of a high luminance level is merely judged a caption but also a caption is judged by further detecting a sharp rate of change of an luminance level which occurs between the black edging applied around a caption and this white caption. Therefore, an image part of a high luminance level is not moved into an image region by expanding an image but only a caption is inserted and therefore it is possible to realize a high quality expansion process in a television with an aspect ratio of 16:9.

Claims

A caption detection device comprising:
means for receiving a luminance signal corresponding to an input image signal which represents an image comprising a caption region in which captions are displayed;

rate of change detecting means for detecting a rate of change of a luminance level of the luminance signal to generate a rate of change signal corresponding to the rate of change;

rate of change judging means for comparing a rate of change level of the rate of change signal with a level of a reference signal inputted separately and for setting a content of a control signal, which represents the comparison result, to valid when the rate of change level exceeds the reference level;

gate means for outputting the luminance signal only if the control signal from the rate of change judging means represents valid; and

caption judging means for comparing a luminance level of the luminance signal outputted from the gate means with a predetermined reference level and for setting a content of a caption signal, which represents the comparison result, to valid when the luminance signal exceeds the reference level.
A caption detection device as claimed in claim 1, further comprising:
storage means for storing the caption signal outputted from the caption judging means; and

control signal generating means for performing a writing control such that the caption signal corresponding to the image of the caption region is written into the storage means and a reading control such that the caption signal is read from the storage means;
wherein the reading control is performed when the caption signal corresponding to a caption insertion region in the image composed by the caption signal is generated.
A caption detection method comprising the steps of:
receiving a luminance signal corresponding to an input image signal which represents an image comprising a caption region in which captions are displayed;

detecting a rate of change of a luminance level of the luminance signal to generate a rate of change signal corresponding to the rate of change;

comparing a rate of change level of the rate of change signal with a level of a reference signal inputted separately and for setting a content of a control signal, which represents the comparison result, to valid when the rate of change level exceeds the reference level;

passing the luminance signal only if the control signal represents valid;

comparing a luminance level of the passed luminance signal with a predetermined reference level and for setting a content of a caption signal, which represents the comparison result, to valid when the passed luminance signal exceeds the reference level.
A caption detection method as claimed in claim 3, further comprising the steps of:
storing the caption signal into a storage means; and

performing a writing control such that the caption signal corresponding to the image of the caption region is written into the storage means and a reading control such that the caption signal is read from the storage means:
wherein the reading control is performed when the caption signal corresponding to a caption insertion region in the image composed by the caption signal is generated.
A video display device, comprising:
a caption detection device as claimed in claim 1 for furnishing a caption signal;

means for inserting a caption at a shifted position into an image in dependence upon the caption signal, to obtain an output image, and

means for displaying the output image.